A bottom-up approach with a clear view of the top:
How human evolutionary psychology can inform adaptive behavior research
Geoffrey F. Miller
School of Cognitive and Computing Sciences
University of Sussex
Falmer, Brighton BN1 9QH, England
(Now at Department of Psychology, University of Nottingham;
gfm@psychology.nottingham.ac.uk)
Peter M. Todd
Department of Psychology
University of Denver
2155 S. Race Street
Denver, CO 80208 USA
ptodd@pstar.psy.du.edu
The Adapted Mind: Evolutionary Psychology and the Generation of Culture. Edited by Jerome H.
Barkow, Leda Cosmides, and John Tooby. New York: Oxford University Press (1992). ISBN 0-19506023-7, xii + 666 pp., $55 hardback.
Introduction
Psychologists have long paid lip service to Darwin, conceding that the human brain did arise
through the course of evolution (for whatever, often unspecified, reason). But the full power of
Darwinian theory is almost never used in day-to-day psychology research. This is peculiar, given
the successful, integrated nature of evolutionary biology, and the typically fragmented and
incomplete visage of modern psychology: one would think that a theory that explains the origin and
maintenance of complex behavioral adaptations across all species (evolution) could inform and
unify the study of human behavior (psychology) just as productively as it does the study of animal
behavior (ethology and comparative cognition). But the emergence of a genuinely evolutionary
psychology of humans (HEP) has been a slow, painful, and quite recent occurrence, marked at last
by the publication of a flagship volume, The Adapted Mind. This work is of great importance not
only for researchers in all branches of human psychology, but also for those in the field of adaptive
behavior as well, because it demonstrates that the operation of even the most sophisticated and
seemingly general behavioral system we know of -- the human brain -- can only be understood as
the conglomeration of a great variety of specific mental mechanisms, each finely tuned to a
particular adaptive problem. Thus, the bottom-up, modular, often evolutionary approach our field
generally takes to the simulation of adaptive behavior (SAB), is just the approach most closely
aligned with the modular, adapted, domain-specific nature of intelligence all the way up to the
human level. As a result, the new HEP provides support for the belief that SAB research is likely
to be the most promising approach to the creation of intelligent adaptive behavior at any level of
sophistication.
Some history
Psychology has not always lacked a unified theoretical foundation. Before Watson's Behaviorism
divorced experimental psychology from the rest of behavioral biology early this century, Darwinian
psychologists such as James Mark Baldwin, William James, George McDougall, Conway Lloyd
Morgan, George Romanes, Herbert Spencer, and indeed Darwin himself, took evolution as the
theoretical framework for analyzing and explaining human and animal behavior (see Richards,
1987). But the Behaviorists (as other schools of psychology that followed) turned to physics as a
paragon of scientific method and mechanistic reductionism, abandoning psychology's natural
affinity with comparative biology as a framework for scientific theory and functionalist analysis.
As a consequence, Behaviorism rejected Darwinian theory along with introspective research
methods and mentalistic concepts, in an attempt to make psychology a positivistic and independent
science.
When mentalistic concepts were revived with the cognitive psychology revolution of the 1950s and
1960s, Darwinian theory remained marginalized. This was a particularly peculiar development,
because the actual overthrow of Behaviorism was instigated by Chomsky's nativist
psycholinguistics (see Pinker, 1994), and by the investigation of innate, domain-specific constraints
on learning by Garcia and Koelling and many others (see Bolles & Beecher, 1988). But for the
next couple of decades, the metaphor of "mind as computer" seemed more promising than the
reality of "mind as evolved biological organ". Psychologists persisted in studying the proximate
mechanisms underlying human behavior without considering their ultimate adaptive functions.
During the 1970s and 1980s, several developments laid the groundwork for a new evolutionary
psychology. New developments in evolutionary theory, from Hamilton's inclusive fitness ideas to
Wilson's conception of sociobiology, paved the way for a much more powerful science of animal
behavior, as well as a more sophisticated and balanced understanding of human evolution (e.g.
Foley, 1987). Field studies of non-human primates revealed surprising homologies to human
behavior and quite sophisticated social and cognitive capacities (e.g. Byrne & Whiten, 1988).
Studies of human cognitive development revealed that infants developed domain-specific
competencies (e.g. face recognition, intuitive physics, social referencing) much earlier and more
spontaneously than standard learning theory could account for (see Hirschfeld & Gelman, 1994).
Artificial intelligence research revealed that real perception and motor control are much more
difficult than previously suspected, especially in comparison to abstract reasoning and problemsolving. Furthermore, almost every computer-simulated (and biological) cognitive process suffers
from problems of combinatorial explosion and the "frame problem", unless guided by "innate"
domain-specific knowledge and constraints on learning. Finally, in the late 1980s, Leda Cosmides
and John Tooby suggested that a new field of "evolutionary psychology" could combine the
evolutionary framework of sociobiology with the psychological sophistication of cognitive science
(Cosmides & Tooby, 1987). They outlined a theoretically sophisticated and empirically productive
method of analyzing human mental capacities as complex biological adaptations, sculpted over
evolutionary time through natural and sexual selection. The field has been burgeoning ever since.
Evolutionary psychologists, clustered at institutions like the University of Michican and the
University of California at Santa Barbara, tend to be renegades from mainstream psychology,
physical anthropology, primatology, and behavioral ecology and biology. Work in this field is
often published in the journals Behavioral and Brain Sciences, Ethology and Sociobiology, and
Human Nature, and the Human Behavior and Evolution Society (HBES) has yearly meetings of
high quality and energy (contact Patrick McKim at pmckim@oboe.aix.calpoly.edu for membership
information and pointers to the HBES electronic mailing list). Evolutionary psychology is in a
difficult position because some biologists and psychologists find it politically correct or convenient
to downplay the relevance of evolutionary theory for human behavior, and because some popular
science writers, feminists, social scientists, and humanists seem to delight in attacking Darwinism
wherever they find it. Nevertheless, evolutionary psychology is attracting support because it is the
only branch of psychology that allows the study of human behavior to be conceptually integrated
into the rest of natural science.
Themes in evolutionary psychology
Biologists expect a functional mesh between the adaptive problems posed by a particular
environment and the design features of particular species of organisms. Evolutionary psychology
posits that humans are no exception: far from having evolved a mythical general-purpose
intelligence that transcends all such specializations, we simply have a different set of speciesspecific mental adaptations evolved to deal with the problems that were posed by our ancestral
environment. The key strategy in evolutionary psychology, then, is to analyze human mental
adaptations based on their adaptive functions under our ancestral conditions. Since the most recent
ancestral environment that could have shaped complex biological adaptations in humans was the
Pleistocene savanna habitat of Africa, evolutionary psychologists think more about the lifestyles of
hunter-gatherers 100,000 years ago, than about the lifestyles of modern humans. Human babies are
born "expecting" to grow up as hunter-gatherers living in migratory tribes with complex social
systems, intense parental and kinship bonds, and fluid sexual relations. Thus, our perceptual,
cognitive, emotional, motivational, and behavioral adaptations are attuned to that ancestral
environment, not to modern conditions. By analyzing the problems of survival and reproduction
that faced our ancestors, and the perceptual cues and behavioral options available under ancestral
conditions, evolutionary psychologists seek insight into the structure of the psychological
adaptations or "Darwinian algorithms" that evolved to solve those problems.
Mental adaptations are not "constraints" on human behavior, that place limits on an otherwise
rational, general-purpose intelligence; no such intelligence is possible, even in principle (see Tooby
and Cosmides's first chapter). Rather, domain-specific mental adaptations are ENABLERS of
behavior that allow competencies and actions that would not otherwise be possible. This "bottomup" view of how evolution constructs minds and behaviors will be familiar to SAB researchers. As
William James and William McDougall argued long ago, humans are unique not because we have
somehow been freed from our instincts, but because we have evolved MORE instincts (mental
adaptations) than other animals, allowing a greater diversity and flexibility of behavior.
Recognizing the importance of these instincts need not lead to a pessimism about human potential:
as Tooby and Cosmides point out, "The human psychological system is immensely flexible as to
outcome: Everything that every individual has ever done in all of human history and prehistory
establishes the minimum boundary of the possible. The maximum, if any, is completely unknown."
(p. 40).
The physical and biological sciences take for granted a COMPATIBILITY principle: chemistry
must be consistent with physics, even if chemistry cannot (yet?) be derived from physics. Mutual
compatibility between sciences leads to conceptual integration. Darwin's central importance in the
history of science lies in his extension of the compatibility requirement to the biological and
psychological realms: evolutionary theory provides the explanatory bridge between physical
simplicity and functional organic and behavioral complexity. Psychology has tried, and failed, to
develop an autonomous "physics of the mind", based on general laws of learning and of cognitive
processing -- laws that are completely uninformed by the principles of selection and adaptation that
apply to other physical and behavioral adaptations in other species. Evolutionary psychology, if
successful, will simply become psychology proper -- a field that is conceptually integrated with the
rest of science, rather than a field seeking a wrong-headed independence.
The book itself
The Adapted Mind comprises 18 chapters from 25 of the best researchers in the field, yielding a
commanding 666 pages of text. Startlingly under-promoted by Oxford University Press, the
volume has been reviewed infrequently and marketed sporadically, but deserves much wider
attention. Jerry Barkow, Leda Cosmides, and John Tooby performed outstanding feats as editors,
actively shaping the diverse contributions so they display a rare unity of vision and purpose, and a
high standard of clarity and quality.
References follow each chapter, and comprehensive author and subject indexes close the book.
The volume's fascination can best be illustrated by a quick skim through the subject index, which
contains a tantalizing diversity of entries such as: adultery, baboons, chastity belts, color vision, ego
defenses, environmental signals (clouds, flowers, water), food sharing, gossip, harems, homicide,
incest, jealousy (morbid), !Kung San, learnability theory, mate-guarding, nepotism, prisoner's
dilemma, rough-and-tumble play, sperm competition, tolerated scrounging, urban design
applications of evolutionary aesthetics, and vomiting. This wealth of topics drives home a key
insight of HEP: that human mental capacities are largely domain-specific adaptations specialized
for dealing with particular kinds of perceptual cues, conceptual contents, and behavioral responses.
There is no short-cut to understand the Mind in general; one must work through each of its
constituent adaptations for solving particular kinds of information-processing tasks, as the authors
in The Adapted Mind set themselves to do.
This functional specialization principle is widely accepted by physiologists for bodily organs and
by psychologists for sensory and motor systems (see Fodor, 1983). HEP simply pushes the
principle to its logical limit, to include the "central processing" underlying cognition, motivation,
and emotion in higher animals. Thus, the book will strike traditional psychologists as having an
unusual structure: there are no chapters on perception in general, or memory in general, or
decision-making in general, or action in general. Each mental adaptation for a particular content
domain is assumed to include perceptual, conceptual, reasoning, decision-making, learning,
memory, and motor control subsystems. In this respect the HEP model of mind resembles Brooks's
subsumption architecture for autonomous robot behavior, with independent layers or modules
guiding each sort of behavior in parallel, interacting only as necessary to avoid behavioral conflict.
Individual chapters
The individual chapters in The Adapted Mind bear varying degrees of applicability for researchers
trying to study or create adaptive behavior in robots or simulated agents. Tooby and Cosmides's
first chapter lays out the arguments for the necessary conceptual integration of the social and
behavioral sciences with the natural sciences, through the medium of evolutionary theory;
researchers in the adaptive behavior field should also be sensitive to this call for integration with
biology in particular, since much of our work to date has been criticized for a lack of familiarity
with the data from the "real" behaving systems we hope to emulate and understand. (This does not
mean that simulations of adaptive behavior must model specific species or neural systems -- more
abstract models, informed by knowledge of biology and the logic of evolution, can be used to
develop useful generalizations ACROSS species.) They view (human) mental mechanisms as the
generators and filters that create and shape the sea of culture we are immersed in, in strict
accordance with the evolved information processing abilities and behavioral predilections of those
mechanisms. Such a view would be a useful starting point for anyone simulating the evolution of
language, culture, or memetic systems.
The second section of the book focuses on the mental mechanisms that have evolved in humans and
other primates to deal with the adaptive problems of exchanging resources between individuals.
Such social exchange algorithms will be of use in any system where individuals can choose to act
so as to benefit another individual, from simple simulations where individuals play the iterated
prisoner's dilemma against each other (Axelrod, 1987), to interacting robots that must cooperate to
accomplish a goal (McFarland, 1994). As Cosmides and Tooby discuss, social exchange
mechanisms must incorporate a variety of design features, including the ability to judge the costs
and benefits of each act of exchange, a means of recognizing other individuals with which
exchanges have taken place, and a way to tell if other individuals are cheating -- that is, accepting a
benefit without subsequently offering a benefit in return. Cheaters must be detected and punished
(or at least shunned), no less for robots cooperating to clean up a shop floor than for hominids
sharing meat after a successful hunt. (Simulations of simpler social systems may be able to show
that some other behavioral mechanisms for social exchange could also evolve and prove adaptive,
which would be an exciting discovery.)
The psychology of choosing mates and holding onto them is discussed in the chapters in the third
section. Since the name of the game in evolution is reproduction, there are very important and
influential evolved mental mechanisms for ensuring that reproductive effort will not be wasted.
Mates must be chosen to maximize the number of viable offspring that are produced, and once
chosen, should be induced to continue to work toward that reproductive goal, rather than heading
off to reproduce with someone else. While reproduction is certainly important in evolutionary
simulations, relatively little work has been done in the adaptive behavior field on the issues of mate
choice or retention (see Collins & Jefferson, 1992, and Todd & Miller, 1993, for exceptions); and
in the case of physical robots, the difficulties of implementing "robot sex" have rendered
reproductive matters pointless.
But the issues covered in this section can be related to any situation in which agents must choose
others to cooperate with toward some fitness-increasing goal. This is true whether fitness is
defined in terms of direct production of simulated offspring, or indirect assessment of "design
success" (including end-user aesthetic and biophilic appeal and other aspects of the robot's
economic niche) and an eventual increase in the number of robots built back at the factory with that
design. For instance, if one type of agent puts more time and energy into a cooperative task than
another type of agent does (as females do in mammalian and avian reproduction), then they will be
a limiting resource that the second agent type will compete for access to and control over (as males
do -- see Wilson and Daly's chapter). These kinds of behavior will require the presence of
information processing mechanisms for assessing time/energy investment and allowing competition
within the agent types and cooperation between them. Furthermore, such choice behaviors can
induce sexual selection and other effects on the course of evolution itself.
In humans and many other species, reproduction is not simply a matter of making babies, but also
of raising and nurturing those offspring at least until they too are reproductively capable.
Significant parental investment is necessary to ensure the survival of human infants to reproductive
age, guided by significant evolved mental mechanisms, as the fourth section on parental care makes
clear. Some of these behavioral adaptations are peculiar to mammalian biology (e.g., Profet's
chapter on pregnancy sickness as a means of reducing the exposure of the developing embryo to
toxins ingested by the mother) or to human birth patterns (e.g., Mann's chapter on maternal
preference for healthier preterm twins). But others are more generally applicable to any situation
where one agent has an interest in fostering the development of another (e.g., an experienced robot
teaching a new robot the best way to navigate through an office environment, to save human
controllers from having to do so).
Fernald, for instance, describes the peculiar features of "motherese" or infant-directed speech that
are specially designed to enhance mother-infant communication: higher pitch for greater salience to
the infant's developing auditory system; exaggerated and continuous pitch contours to enhance
voice tracking; and highly redundant, small-repertoire vocalizations for increased signal detection.
These design features not only initiate the communication of emotional and motivational states
between parent and child, but also can serve to help demarcate the symbolically important
linguistic elements of speech, focusing attention on individual words and thereby aiding in further
language learning. Consideration of design features such as these (including setting up situations
where teaching and learning can co-evolve) will be important for enhancing communication to
robots or animats whose behaviors and "brains" are still developing, and for leading them to more
effective learning by highlighting the aspects of the environment that should be internalized first.
Boulton and Smith consider a developmental process that typically happens independently of
adults: rough-and-tumble play. This kind of activity, it is hypothesized, may help children (and
young animals) to learn locomotor skills that will be necessary in adulthood, for fighting, hunting,
or predator avoidance. Some form of "play" behavior may also be very useful to give more or less
fragile robots the opportunity to practice and fine-tune motor control programs that may be crucial
to later survival (e.g. escaping predators or other destructive forces -- see Miller & Cliff, 1994) in
an initially benign and "playful" setting, removed from the consequences of making costly mistakes
(e.g. being crushed by an overly-curious child).
The fifth section, on language and perception, begins with Pinker and Bloom's carefully reasoned
and convincing argument (reprinted from Behavioral and Brain Sciences) for the evolution of
human language by natural selection, rather than as a "spandrel" or non-functional quirk. They
discuss the design features and adaptive advantages of language, and provide criticism of the antiadaptationist camp that will be valuable for anyone making selection-based explanations of evolved
behavioral systems. And their paper will be inspirational in reminding SAB researchers that
communication systems based on arbitrary symbolic representation and complex grammars CAN
evolve (rather than just appear), even if current simulations are still far from this goal.
Shepard presents a theory for the human perception of color along three independent dimensions,
stating that this design is used to achieve the important adaptive function of color constancy -keeping objects identifiable by their color across widely-varying illumination conditions (bright
sun, cloudy days, moonlight, etc.). Shepard holds that this aspect of the visual system, in common
with other perceptual and cognitive abilities like the representation of motion of objects in space, is
strongly attuned to structure in the world. In particular, he proposes that those characteristics of the
world that have been present over the greatest amount of evolutionary time (e.g., light coming from
the sun, and space being three dimensional), will be most deeply internalized. While this proposal
for human color perception has yet to be proven (it is problematic that some primates have only
two types of color-sensitive cones, and thus perhaps a different color perception dimensionality),
the general results from Shepard's research program remain applicable to the design of the "deep"
underlying perceptual and cognitive abilities of any agents inhabiting our physical world.
Shepard's work also presents strong arguments for the use of internal representations in cognition,
and so should be read by anyone pondering intelligence without representation (see e.g. Shepard,
1987).
Finally in this section, Silverman and Eals describe their experiments showing the existence of
different mechanisms for spatial cognition, suggested by an analysis of the different spatial
adaptive problems that would have faced male hunters of mobile prey and female gatherers of
stationary plants in Pleistocene Africa. Their work points up the importance once again of
carefully analyzing the adaptive problems faced by a behaving agent, and then considering the taskspecific, rather than domain-general, solutions that can address those problems.
The two chapters in section 6, on environment aesthetics, describe how humans gain information
about their surroundings, and use that information to make judgments between alternative
environments. Orians and Heerwagen break the problem of habitat selection into three stages: an
initial rapid decision whether or not to explore a newly-encountered landscape; a longer period of
exploration of possibly suitable landscapes; and a final decision to stay in a region for some length
of time or to move on in search of greener pastures (or savannas, in keeping with our Pleistocene
origins). Cognition at each of these stages is sensitive to cues in the environment reflecting
information about varying time-spans. These include transitory changes in weather, longer-term
seasonal shifts, and nearly-permanent geographical features. Kaplan describes responses to
environments in terms of the prospects for new information they afford to knowledge-seeking
individuals: environments where exploration seems possible and productive are more highly valued
than those where it seems easy to get lost, or where endless, featureless expanses preclude
discovering anything new. Both sets of authors emphasize that the rapid, affect-laden nature of
human aesthetic responses to different environments indicates the important adaptive functions of
these responses. Making quick decisions about where to seek refuge or where to head to look for
prey could be very advantageous in the shifting savanna. Furthermore, imbuing the reactions to
landscapes with strong emotional feelings helps to organize thought and action for the appropriate
responses (fleeing, exploring, settling, etc.). The results of these research programs could be useful
in building exploratory behavior, environment navigation, or "nesting site" choice into agents and
robots, and the ideas about the role of affect in cognition and judgment are important for the design
and understanding of any motivational system.
The penultimate chapter in the book, by Nesse and Lloyd, is about the evolution of psychodynamic
mechanisms including repression, neurosis, transference, and guilt -- the types of processes focused
on by psychoanalysis. This would seem a topic of singularly little interest to researchers in the
SAB field (at least while at work). But in fact, the emphasis here is on intrapsychic conflict, how
differences of motivation and decision between different cognitive modules within one individual
can be resolved. As such, this topic is very germane to action selection algorithms and interacting
motivational systems. One of the prime means of alleviating conflicting opinions between modules
is to repress the outputs of one of the modules -- that is, to deceive oneself about one's own
cognitions. Here, perceiving one's own internal states veridically is not necessarily the most
adaptive course of behavior, just as perceiving the external world veridically is not always adaptive,
either: in both cases, what is important is to process the relevant fitness affordances -- the fitnessaffecting cues in the internal or external environment -- even if they contradict or mask the
objective "facts" (like the actual output of an internal motivational module). Such self-deception
may be particularly adaptive when trying to deceive others; if one's underlying motivations are
hidden even from one's own conscious awareness, they will be that much more difficult for other
competing individuals to detect and exploit. While simulations of interacting agents are certainly
not yet at the level of sophistication where self-deception (or even deception of others?) is called
for, these ideas should be kept in mind for the future.
How HEP can inform SAB
The research program of evolutionary psychology helps to elucidate the adaptive problems and
information-processing solutions, such as cheater-detection (chapter 3), that emerged over the
course of human evolution. Researchers in SAB can use these ongoing new results from HEP most
directly in two ways. As designers of new adaptively-behaving systems, we can incorporate the
functions and behavioral mechanisms uncovered by evolutionary psychology into our robots or
animats; and alternatively, as shepherds of simulated evolutionary design processes, we can use the
results of evolutionary psychology to help us look for and understand the adaptive informationprocessing components in our evolved neural network or genetic programming systems. In both
cases, it is important to remember that the mechanisms involved need not be as complex as those
found in the brains of humans (or other living organisms) -- for instance, animats looking for
cheating social partners need not incorporate things like voice stress analysis or checking for shifty
eyes, but rather can rely on the simpler cues and fitness affordances present in their simplified
worlds. Thus, the ultimate adaptive functions of behaviors will map from the human (or animal) to
the animat (or robot) case, even if the proximate mechanisms do not.
The new evolutionary psychology view of mental origins and functions puts the simulation of
adaptive behavior field on much firmer ground, and in a much more important position. SAB
researchers often confront the objection that, though the behavior of simple animals can be
modeled as the outcome of domain-specific mechanisms adapted to particular environmental
problems, surely there will be a need for more domain-general reasoning and learning capacities as
we climb the ladder of complexity from artificial life to artificial intelligence. Familiarity with
human HEP gives us a powerful defense against this objection, and a strong justification for
viewing mental evolution as "adaptations all the way up". Tearing down the scientifically
unjustifiable (but politically convenient) wall between human and animal cognition gives the SAB
field more relevance to human psychology than even its most ardent supporters might have
suspected. Additionally, in highlighting the continuity between animal and human behavior,
knowing about human evolutionary psychology may also help SAB researchers on a more personal
level, by making SAB work more clearly relevant to human issues, and helping us better
understand the real human organisms we work with.
How SAB can inform HEP
Cognitive psychology has dominated human psychology for the last twenty years, and the computer
metaphor has always dominated cognitive psychology. Because of this, some evolutionary
psychologists have found it tactically useful to adopt cognitivist and computational language in
discussing mental adaptations, to make Darwinian theory more comprehensible and palatable to
mainstream cognitive psychology. Indeed, Cosmides and Tooby often refer to mental adaptations
as "Darwinian algorithms" that take in specific perceptual cues, transform them into internal
representations of environmental situations, process those representations according to domainspecific inference rules, and output certain motor patterns in consequence. This cognitivism can be
off-putting to SAB researchers who are skeptical of the entire cognitive science tradition, and who
prefer to think of mental and behavioral adaptations using other models, such as dynamical systems
theory (see e.g. Port & Van Gelder, in press), neural network theory, or classifier systems. It would
be unfortunate if HEP became conflated with a particular cognitivist metaphor for how mental
adaptations operate, because HEP can easily work with other metaphors more palatable to the SAB
community. For example, Miller and Freyd (1993) have outlined how evolutionary psychology,
dynamical systems models of cognition, and a broad notion of internal representation are mutually
compatible. SAB research can help guide HEP by supplying a broader palette of metaphors for
cognition, making HEP more adaptive in the face of changes in computational fashion in the ranks
of old-fashioned psychology.
Conclusion: Towards a universal science of behavioral evolution
Evolutionary psychology as developed in The Adapted Mind offers a theoretical framework and
experimental methodology for studying the evolutionary origins, adaptive functions, and proximate
mechanisms of human behavior. Research on the simulation of adaptive behavior as developed in
this journal offers computational methods for studying the evolution, functions, and mechanisms of
intelligent behavior in general. Thus, HEP and SAB work share a congruence of theoretical
interests and a complementarity of empirical methods.
However, there remains a large gap between the kinds of behaviors modeled in SAB and the kinds
of behaviors investigated by HEP. This gap will probably take generations to close, but until then,
it is important for each field to respect and learn from the other. Human evolutionary psychology
offers a vision of the human mind that is radically different from the domain-general cognitive
engine postulated by traditional psychology, or the culturally-constructed semiotic network
conjectured in contemporary social science and the humanities, or the black box full of
individualistic beliefs and desires assumed by the crypto-creationist portions of the public. To the
extent that SAB researchers have tacitly absorbed one of these latter three models from our current
largely anti-Darwinian culture, and to the extent that these models bias our understanding of what
constitutes complex adaptive behavior, we will aim in exactly the wrong direction in our research.
Reading The Adapted Mind will be a useful prophylactic against our field repeating the mistakes
that psychology has been making for the last 100 years, and will provide inspiration to help forge
the unified science of the evolution of adaptive behavior during the coming century.
References
Axelrod, R. (1987). The evolution of strategies in the iterated Prisoner's Dilemma. In L. D. Davis
(Ed.), Genetic algorithms and simulated annealing (pp. 32-41). Los Altos, CA: Morgan Kaufmann.
Bolles, R. C., and Beecher, M. D. (Eds.) (1988). Evolution and learning. Hillsdale, NJ: Erlbaum.
Byrne, R., and Whiten, A. (Eds.) (1988). Machiavellian intelligence: Social expertise and the
evolution of intellect in monkeys, apes, and humans. Oxford, UK: Oxford University Press.
Collins, R. J., and Jefferson, D. R. (1992). The evolution of sexual selection and female choice. In
F. J. Varela and P. Bourgine (Eds.), Toward a practice of autonomous systems: Proceedings of the
First European Conference on Artificial Life (pp. 327-336). Cambridge, MA: MIT Press/Bradford
Books.
Cosmides, L., and Tooby, J. (1987). From evolution to behavior: Evolutionary psychology as the
missing link. In J. Dupre (Ed.), The latest on the best: Essays on evolution and optimality (pp. 277306). Cambridge, MA: MIT Press/Bradford Books.
Fodor, J. A. (1983). Modularity of mind. Cambridge, MA: MIT Press.
Foley, R. (1987). Another unique species: Patterns in human evolutionary ecology. Harlow, UK:
Longman Scientific & Technical.
Hirschfeld, L. A., and Gelman, S. A. (Eds.) (1994). Mapping the mind: Domain specificity in
cognition and culture. Cambridge, UK: Cambridge University Press.
McFarland, D. (1994). Towards robot cooperation. In D. Cliff, P. Husbands, J.-A. Meyer, and S.
W. Wilson (Eds.), From animals to animats 3: Proceedings of the Third International Conference
on Simulation of Adaptive Behavior (pp. 440-444). Cambridge, MA: MIT Press/Bradford Books.
Miller, G. F., and Cliff, D. (1994). Protean behavior in dynamic games: Arguments for the coevolution of pursuit-evasion tactics. In D. Cliff, P. Husbands, J.-A. Meyer, and S. W. Wilson
(Eds.), From animals to animats 3: Proceedings of the Third International Conference on
Simulation of Adaptive Behavior (pp. 411-420). Cambridge, MA: MIT Press/Bradford Books.
Miller, G. F., and Freyd, J. J. (1993). Dynamic mental representations of animate motion: The
interplay among evolutionary, cognitive, and behavioral dynamics. Cognitive Science Research
Paper CSRP-290, University of Sussex.
Pinker, S. (1994). The language instinct. New York: Basic Books.
Port, R. F. and Van Gelder, T. (Eds.). (in press). Mind as motion: Dynamics, behavior, and
cognition. Cambridge, MA: MIT Press.
Richards, R. J. (1987). Darwin and the emergence of evolutionary theories of mind and behavior.
Chicago: University of Chicago Press.
Shepard, R. N. (1987). Evolution of a mesh between principles of the mind and regularities of the
world. In J. Dupre (Ed.), The latest on the best: Essays on evolution and optimality (pp. 251-275).
Cambridge, MA: MIT Press/Bradford Books.
Todd, P. M., and Miller, G. F. (1993). Parental guidance suggested: How parental imprinting
evolves through sexual selection as an adaptive learning mechanism. Adaptive Behavior, 2(1), 547.